Stress and coping processes affect biomarkers such as blood pressure, cholesterol fractions, and cortisol in laboratory settings, and self-reported health in field settings. However, establishing the stress-illness link has been very difficult (Cohen et al., 2007), in part because many epidemiological studies of stress have single measures of life events which, at best, are only weak predictors of long-term health. At the Normative Aging Study, we have been collecting stress and coping data for 20+ years, which provides a unique opportunity to examine how patterns of stress over the long term (e.g., chronic vs. intermittent vs. low) affect changes in biomarkers, the development of cardiovascular disease, and mortality, as well as long-term mental health trajectories. Our theoretical model proposes that individuals can decrease vulnerability to disease and promote optimal aging through judicious management of stress, including appraisal and coping processes. The Normative Aging Study (NAS) is a large panel study of 2,280 men aged 21 to 80 at study onset in the 1960s. In 1985, we began studying life events, and in 1989 began collecting, life events, hassles and coping measures in conjunction with the triennial biomedical exams. In 2001, we added daily stressors using diary studies, and included NAS wives. In 2008 we added cortisol to the daily diary studies. The proposed additional data collection in the NAS will allow us to utilize a longitudinal burst design to examine the interplay among chronic stress, daily stress, personality, and biomarkers such as cortisol diurnal rhythms and alpha amylase in older men and women. We will also examine change in cortisol reactivity over time, thought to be a key factor modulating the rate of aging. The first specific aim will be to model patterns of change in stress and coping processes from middle-age to late life, using growth mixture models. Second, we will examine the impact of these different patterns on changes in biomarkers (e.g., blood pressure, cholesterol fractions), as well as the development of cardiovascular disease and mortality. Using coupling analyses, we hypothesize that hassles will better predict change in the more proximal biomarkers, while life event patterns will be better predictors of long-term health outcomes. In contrast, positive coping will promote the maintenance of good mental health as well as HDL levels. The third specific aim will examine the impact of chronic stress, personality, gender, and aging on cortisol reactivity.